On 12/3/19 4:16 PM, Thomas Schwinge wrote:
On 2019-11-15T20:11:29+0100, Tobias Burnus <tob...@codesourcery.com> wrote:
* Make no_create.c effective by adding 'has_firstprivate = true;' to
target.c.*
(* If one tries to access c or e in the no_create-3.{c,f90} run-time
test case, plugin-nvidia rightly complains (illegal memory access),
using the created 'b' or 'd' works as tested by the test case.
So that's specifically what you fixed above, or is that another problem?
Well, that was one way of manually testing that it really worked for
not-mapped variables w/o creating them (i.e. verifying that "no_create"
didn't just act as "present"). – Manual as that's not that simple to
code in the test suite (shared memory, exact wording for dg-output etc.)
— However, I think it can be done using '#include <openacc.h>' / "use
openacc", #if !ACC_MEM_SHARED, and calling acc_is_present (passing
either "sizeof()" or a simple "1" as "len" argument); hence, I will try
this next version of the patch.
I'm willing to accept that patch as-is, unless Jakub has any further comments
at this point. […]
With these items considered/addressed as you feel comfortable, this is OK for
trunk.
Tobias
PS: I will have a closer look tomorrow at the your new test cases and
comments.
